All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH RESEND v2] x86: warn on native_io_delay() before DMI scan
@ 2021-10-17 19:25 Mateusz Jończyk
  2021-12-23 11:05 ` Borislav Petkov
  0 siblings, 1 reply; 2+ messages in thread
From: Mateusz Jończyk @ 2021-10-17 19:25 UTC (permalink / raw)
  To: linux-kernel, x86
  Cc: Mateusz Jończyk, Thomas Gleixner, Ingo Molnar, Borislav Petkov

Writing to IO port 0x80 in native_io_delay() causes system lockups on
some laptops. This is handled by io_delay_init(), which scans DMI info
and changes io_delay_type on these laptops.

Therefore, calling native_io_delay() / outb_p() / inb_p() / etc. before
io_delay_init() may cause problems there and may constitute a kernel
bug. Warn if this happens.

Signed-off-by: Mateusz Jończyk <mat.jonczyk@o2.pl>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Borislav Petkov <bp@alien8.de>
---

v2: make io_delay_dmi_scanned static

 arch/x86/kernel/io_delay.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/arch/x86/kernel/io_delay.c b/arch/x86/kernel/io_delay.c
index fdb6506ceaaa..d6deca551a5c 100644
--- a/arch/x86/kernel/io_delay.c
+++ b/arch/x86/kernel/io_delay.c
@@ -29,6 +29,7 @@
 #endif
 
 int io_delay_type __read_mostly = DEFAULT_IO_DELAY_TYPE;
+static int io_delay_dmi_scanned __read_mostly;
 
 static int __initdata io_delay_override;
 
@@ -37,6 +38,8 @@ static int __initdata io_delay_override;
  */
 void native_io_delay(void)
 {
+	WARN_ON_ONCE(!io_delay_dmi_scanned);
+
 	switch (io_delay_type) {
 	default:
 	case IO_DELAY_TYPE_0X80:
@@ -123,6 +126,8 @@ void __init io_delay_init(void)
 {
 	if (!io_delay_override)
 		dmi_check_system(io_delay_0xed_port_dmi_table);
+
+	io_delay_dmi_scanned = 1;
 }
 
 static int __init io_delay_param(char *s)
-- 
2.25.1


^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [PATCH RESEND v2] x86: warn on native_io_delay() before DMI scan
  2021-10-17 19:25 [PATCH RESEND v2] x86: warn on native_io_delay() before DMI scan Mateusz Jończyk
@ 2021-12-23 11:05 ` Borislav Petkov
  0 siblings, 0 replies; 2+ messages in thread
From: Borislav Petkov @ 2021-12-23 11:05 UTC (permalink / raw)
  To: Mateusz Jończyk; +Cc: linux-kernel, x86, Thomas Gleixner, Ingo Molnar

On Sun, Oct 17, 2021 at 09:25:58PM +0200, Mateusz Jończyk wrote:
> Writing to IO port 0x80 in native_io_delay() causes system lockups on
> some laptops. This is handled by io_delay_init(), which scans DMI info
> and changes io_delay_type on these laptops.
> 
> Therefore, calling native_io_delay() / outb_p() / inb_p() / etc. before
> io_delay_init() may cause problems there and may constitute a kernel
> bug. Warn if this happens.

When/how does this happen?

io_delay_init() gets called pretty early in setup_arch()...

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2021-12-23 11:05 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-10-17 19:25 [PATCH RESEND v2] x86: warn on native_io_delay() before DMI scan Mateusz Jończyk
2021-12-23 11:05 ` Borislav Petkov

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.